More than 1 million Americans die each year from heart and lung diseases. They account for 57 percent of all deaths and cost more than $120 billion each year, or more than 25 percent of the total economic cost of illness and premature death. Since physicians are known to be gatekeepers to the costly treatment and technology associated with cardiopulmonary diseases, it is important to disentangle the factors influencing their everyday clinical decision-making. Moreover, the known variability in prevalence rates for cardiopulmonary morbidity and mortality may be partially explained by differences in decision-making (especially with respect to diagnosis and treatment recommendations). Physicians' evaluations of patients' presenting symptoms are generally thought to be relatively free of extraneous influences and guided by biomedical criteria. However, a great deal of discretion is thought to intrude in the clinical decision-making (CDM) process. Prior research suggests that patient sociodemographics, personal characteristics of physicians and different practice settings influence the decisions reached with respect to diagnosis, recommendations for treatment and referral and hospitalization. Despite extensive research on this topic, most studies to date have been either narrow in scope, or limited by methodological difficulties. We still await a properly designed and executed study, on a sufficiently large and representative sample of medical practitioners, which is able to describe and explain the influence of non-biomedical factors on CDM. Using a factorial design, the proposed research by a multidisciplinary team will systematically examine the independent and joint influences of patient, physician, and practice-setting variables on the CDM process with respect to 3 cardiopulmonary situations. It will identify the costs of diagnostic and therapeutic decisions made by a sample (n=192) of internists and family or general practitioners in the Greater Boston area in response to 48 well-designed and professionally-acted patient vignettes concerning chest pain, chronic dyspnea and a general medical examination.